Answers are grounded in this paper and cite their sources.
Mountainous multi-UAV monitoring couples task assignment with terrain-constrained routing, wind-dependent costs, communication conditions, and dynamic execution. We present SENTRA, a hierarchical framework that integrates sensor-compatible task assignment and sequencing, segment-cost feedback, QLPSO reference-path generation, and Adaptive DWA rolling trajectory adjustment. Three illustrative scenarios with 20, 25, and 40 tasks on a common DEM completed all initialized tasks under the implemented simulator checks. In a paired framework comparison comprising nine newly regenerated scenario instances at three matched nominal configuration levels, the coupled pipeline achieved mean task-completion rates of 90.00%, 76.00%, and 70.00%, compared with 86.67%, 70.67%, and 68.33% for a sequential no-feedback baseline, together with lower mean path length and estimated energy. A ten-seed action-selection ablation on a fixed path segment gave QLPSO the lowest mean cost among the tested fixed, random, and cyclic parameter schedules. In event-recovery tests, all blockage and localized-wind replacement segments were accepted. Blockage recovery achieved 86% full-fleet task completion, while localized-wind recovery reduced mean cumulative wind risk from 7.126 to 0.086; the target UAV completed its task in all ten trials, and full-fleet completion was 88%. These results characterize the integrated planning and segment-level recovery behavior of SENTRA under the tested kinematic simulation settings.
No takes yet. Share an insight, caveat, or question.
Li et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: